rest-example using falcon framework

The application was built in python using memcached, rabbitmq, celery, falcon, gunicorn, and meinheld. There is no build script. The example was built with scalability in mind. There can be multiple frontends and backends running, allowing the code to be distributed, along with caching of results to reduce latency.

install memcached

apt-get install memcached or yum install memcached

download rabbitmq-server and install according to the diretions here

https://www.rabbitmq.com/download.html

if you want to see the rabbitmq server in action

sudo rabbitmq-plugins enable rabbitmq_management

install the miniconda distribution

wget https://repo.continuum.io/miniconda/Miniconda2-latest-Linux-x86_64.sh bash Miniconda2-latest-Linux-x86_64.sh

answer yes to modifying the path in your .bashrc file

source your modified .bashrc file

source ~/.bashrc

execute the folloing conda commands to install the following packages

conda install gunicorn conda install python-memcached conda install --channel https://conda.anaconda.org/axiom-data-science falcon conda install --channel https://conda.anaconda.org/axiom-data-science celery

if python pip not installed

wget https://bootstrap.pypa.io/get-pip.py python get-pip.py

newest meinheld is not in conda, so we pip install it

pip install meinheld

start the celery backends

./runcelery

start the wsgi server

./runserver

unit testing

unittest.py

rest service is hosted on

http://localhost:8000/fibonacci/x

where x is the integer argument for the series to be returned